| { | |
| "best_metric": 4.0290679931640625, | |
| "best_model_checkpoint": "./models/gpt_trinity_test/checkpoint-84000", | |
| "epoch": 4.0, | |
| "global_step": 84736, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 3.533160254897333e-06, | |
| "loss": 3.5765, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "eval_loss": 4.124706745147705, | |
| "eval_runtime": 5.5178, | |
| "eval_samples_per_second": 77.749, | |
| "eval_steps_per_second": 38.965, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 7.073400991267406e-06, | |
| "loss": 3.19, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "eval_loss": 4.057796955108643, | |
| "eval_runtime": 5.5495, | |
| "eval_samples_per_second": 77.305, | |
| "eval_steps_per_second": 38.742, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 1.061364172763748e-05, | |
| "loss": 3.1177, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "eval_loss": 4.070754051208496, | |
| "eval_runtime": 5.5223, | |
| "eval_samples_per_second": 77.685, | |
| "eval_steps_per_second": 38.933, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 1.4153882464007553e-05, | |
| "loss": 3.1116, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "eval_loss": 4.065377712249756, | |
| "eval_runtime": 5.5251, | |
| "eval_samples_per_second": 77.646, | |
| "eval_steps_per_second": 38.913, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 1.7694123200377628e-05, | |
| "loss": 3.0777, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "eval_loss": 4.085710525512695, | |
| "eval_runtime": 5.5077, | |
| "eval_samples_per_second": 77.89, | |
| "eval_steps_per_second": 39.036, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 2.12343639367477e-05, | |
| "loss": 3.1105, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "eval_loss": 4.112701416015625, | |
| "eval_runtime": 5.4987, | |
| "eval_samples_per_second": 78.018, | |
| "eval_steps_per_second": 39.1, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 2.4771064432381404e-05, | |
| "loss": 3.1018, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "eval_loss": 4.141016483306885, | |
| "eval_runtime": 5.5233, | |
| "eval_samples_per_second": 77.671, | |
| "eval_steps_per_second": 38.926, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 2.8311305168751475e-05, | |
| "loss": 3.0728, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "eval_loss": 4.183428764343262, | |
| "eval_runtime": 5.5153, | |
| "eval_samples_per_second": 77.783, | |
| "eval_steps_per_second": 38.982, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 2.9794261886653903e-05, | |
| "loss": 3.1248, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "eval_loss": 4.205781936645508, | |
| "eval_runtime": 5.4948, | |
| "eval_samples_per_second": 78.073, | |
| "eval_steps_per_second": 39.128, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 2.9400881172799034e-05, | |
| "loss": 3.1035, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "eval_loss": 4.20481538772583, | |
| "eval_runtime": 5.5171, | |
| "eval_samples_per_second": 77.758, | |
| "eval_steps_per_second": 38.97, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 2.9007893839658023e-05, | |
| "loss": 3.0943, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "eval_loss": 4.189157009124756, | |
| "eval_runtime": 5.5205, | |
| "eval_samples_per_second": 77.71, | |
| "eval_steps_per_second": 38.946, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 2.8614513125803154e-05, | |
| "loss": 3.0724, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "eval_loss": 4.206292629241943, | |
| "eval_runtime": 5.5117, | |
| "eval_samples_per_second": 77.834, | |
| "eval_steps_per_second": 39.008, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 2.822113241194828e-05, | |
| "loss": 3.0517, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "eval_loss": 4.19225549697876, | |
| "eval_runtime": 5.5111, | |
| "eval_samples_per_second": 77.843, | |
| "eval_steps_per_second": 39.012, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 2.782814507880727e-05, | |
| "loss": 3.0372, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "eval_loss": 4.2112298011779785, | |
| "eval_runtime": 5.4935, | |
| "eval_samples_per_second": 78.093, | |
| "eval_steps_per_second": 39.138, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 2.7434764364952404e-05, | |
| "loss": 3.0235, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "eval_loss": 4.204270362854004, | |
| "eval_runtime": 5.5028, | |
| "eval_samples_per_second": 77.961, | |
| "eval_steps_per_second": 39.071, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 2.704177703181139e-05, | |
| "loss": 3.0329, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "eval_loss": 4.163009166717529, | |
| "eval_runtime": 5.5278, | |
| "eval_samples_per_second": 77.608, | |
| "eval_steps_per_second": 38.894, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 2.664839631795652e-05, | |
| "loss": 3.0171, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_loss": 4.163072109222412, | |
| "eval_runtime": 5.4985, | |
| "eval_samples_per_second": 78.021, | |
| "eval_steps_per_second": 39.102, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 2.625501560410165e-05, | |
| "loss": 2.9997, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "eval_loss": 4.156316757202148, | |
| "eval_runtime": 5.5028, | |
| "eval_samples_per_second": 77.961, | |
| "eval_steps_per_second": 39.071, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 2.5861634890246782e-05, | |
| "loss": 2.9913, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "eval_loss": 4.161618232727051, | |
| "eval_runtime": 5.5279, | |
| "eval_samples_per_second": 77.607, | |
| "eval_steps_per_second": 38.894, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 2.546904093781962e-05, | |
| "loss": 2.9579, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "eval_loss": 4.149376392364502, | |
| "eval_runtime": 5.5138, | |
| "eval_samples_per_second": 77.805, | |
| "eval_steps_per_second": 38.993, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 2.507605360467861e-05, | |
| "loss": 2.9576, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "eval_loss": 4.136674880981445, | |
| "eval_runtime": 5.5162, | |
| "eval_samples_per_second": 77.77, | |
| "eval_steps_per_second": 38.976, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 2.468267289082374e-05, | |
| "loss": 2.7461, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "eval_loss": 4.15926456451416, | |
| "eval_runtime": 5.5064, | |
| "eval_samples_per_second": 77.909, | |
| "eval_steps_per_second": 39.045, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 2.4289292176968872e-05, | |
| "loss": 2.7637, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "eval_loss": 4.1452813148498535, | |
| "eval_runtime": 5.5159, | |
| "eval_samples_per_second": 77.775, | |
| "eval_steps_per_second": 38.978, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 2.3895911463114e-05, | |
| "loss": 2.741, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "eval_loss": 4.162442207336426, | |
| "eval_runtime": 5.5028, | |
| "eval_samples_per_second": 77.961, | |
| "eval_steps_per_second": 39.071, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 2.3502530749259134e-05, | |
| "loss": 2.7514, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "eval_loss": 4.135677814483643, | |
| "eval_runtime": 5.525, | |
| "eval_samples_per_second": 77.647, | |
| "eval_steps_per_second": 38.914, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 2.3109150035404265e-05, | |
| "loss": 2.755, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "eval_loss": 4.152438163757324, | |
| "eval_runtime": 5.5218, | |
| "eval_samples_per_second": 77.692, | |
| "eval_steps_per_second": 38.937, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 2.2715769321549396e-05, | |
| "loss": 2.7365, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "eval_loss": 4.139864444732666, | |
| "eval_runtime": 5.5425, | |
| "eval_samples_per_second": 77.401, | |
| "eval_steps_per_second": 38.791, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 2.232278198840838e-05, | |
| "loss": 2.7356, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "eval_loss": 4.1285014152526855, | |
| "eval_runtime": 5.5006, | |
| "eval_samples_per_second": 77.991, | |
| "eval_steps_per_second": 39.087, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 2.1929401274553512e-05, | |
| "loss": 2.7386, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "eval_loss": 4.128605365753174, | |
| "eval_runtime": 5.5111, | |
| "eval_samples_per_second": 77.843, | |
| "eval_steps_per_second": 39.012, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 2.15364139414125e-05, | |
| "loss": 2.7489, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "eval_loss": 4.123142719268799, | |
| "eval_runtime": 5.5126, | |
| "eval_samples_per_second": 77.821, | |
| "eval_steps_per_second": 39.001, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 2.114303322755763e-05, | |
| "loss": 2.7518, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "eval_loss": 4.110356330871582, | |
| "eval_runtime": 5.5003, | |
| "eval_samples_per_second": 77.995, | |
| "eval_steps_per_second": 39.089, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 2.0750045894416617e-05, | |
| "loss": 2.7317, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "eval_loss": 4.120232582092285, | |
| "eval_runtime": 5.4956, | |
| "eval_samples_per_second": 78.063, | |
| "eval_steps_per_second": 39.122, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 2.0356665180561748e-05, | |
| "loss": 2.7378, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "eval_loss": 4.113248825073242, | |
| "eval_runtime": 5.4916, | |
| "eval_samples_per_second": 78.119, | |
| "eval_steps_per_second": 39.151, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 1.9963677847420733e-05, | |
| "loss": 2.7309, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "eval_loss": 4.104722499847412, | |
| "eval_runtime": 5.5185, | |
| "eval_samples_per_second": 77.739, | |
| "eval_steps_per_second": 38.96, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "learning_rate": 1.9570297133565864e-05, | |
| "loss": 2.7791, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "eval_loss": 4.09762716293335, | |
| "eval_runtime": 5.5051, | |
| "eval_samples_per_second": 77.928, | |
| "eval_steps_per_second": 39.055, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "learning_rate": 1.9176916419710998e-05, | |
| "loss": 2.7427, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "eval_loss": 4.087375164031982, | |
| "eval_runtime": 5.502, | |
| "eval_samples_per_second": 77.972, | |
| "eval_steps_per_second": 39.077, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 1.8784322467283838e-05, | |
| "loss": 2.7184, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "eval_loss": 4.0953216552734375, | |
| "eval_runtime": 5.4993, | |
| "eval_samples_per_second": 78.01, | |
| "eval_steps_per_second": 39.096, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 1.8390941753428972e-05, | |
| "loss": 2.7107, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "eval_loss": 4.09627103805542, | |
| "eval_runtime": 5.5281, | |
| "eval_samples_per_second": 77.604, | |
| "eval_steps_per_second": 38.892, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "learning_rate": 1.7997561039574103e-05, | |
| "loss": 2.7122, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "eval_loss": 4.0841288566589355, | |
| "eval_runtime": 5.5134, | |
| "eval_samples_per_second": 77.811, | |
| "eval_steps_per_second": 38.996, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 1.760418032571923e-05, | |
| "loss": 2.7172, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "eval_loss": 4.085172176361084, | |
| "eval_runtime": 5.5109, | |
| "eval_samples_per_second": 77.845, | |
| "eval_steps_per_second": 39.013, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "learning_rate": 1.721119299257822e-05, | |
| "loss": 2.7126, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "eval_loss": 4.063177108764648, | |
| "eval_runtime": 5.484, | |
| "eval_samples_per_second": 78.227, | |
| "eval_steps_per_second": 39.205, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "learning_rate": 1.681781227872335e-05, | |
| "loss": 2.7063, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "eval_loss": 4.064344882965088, | |
| "eval_runtime": 5.5064, | |
| "eval_samples_per_second": 77.909, | |
| "eval_steps_per_second": 39.045, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 1.642443156486848e-05, | |
| "loss": 2.5311, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "eval_loss": 4.084793567657471, | |
| "eval_runtime": 5.5081, | |
| "eval_samples_per_second": 77.885, | |
| "eval_steps_per_second": 39.033, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 1.6031444231727466e-05, | |
| "loss": 2.4496, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "eval_loss": 4.094330310821533, | |
| "eval_runtime": 5.4892, | |
| "eval_samples_per_second": 78.153, | |
| "eval_steps_per_second": 39.167, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 1.563845689858645e-05, | |
| "loss": 2.4597, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "eval_loss": 4.079880714416504, | |
| "eval_runtime": 5.4837, | |
| "eval_samples_per_second": 78.232, | |
| "eval_steps_per_second": 39.207, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "learning_rate": 1.5245076184731584e-05, | |
| "loss": 2.4472, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "eval_loss": 4.080161094665527, | |
| "eval_runtime": 5.4997, | |
| "eval_samples_per_second": 78.005, | |
| "eval_steps_per_second": 39.093, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "learning_rate": 1.4851695470876715e-05, | |
| "loss": 2.4628, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "eval_loss": 4.088040828704834, | |
| "eval_runtime": 5.492, | |
| "eval_samples_per_second": 78.114, | |
| "eval_steps_per_second": 39.148, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "learning_rate": 1.4458314757021846e-05, | |
| "loss": 2.4508, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "eval_loss": 4.079111576080322, | |
| "eval_runtime": 5.4843, | |
| "eval_samples_per_second": 78.224, | |
| "eval_steps_per_second": 39.203, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "learning_rate": 1.4065327423880831e-05, | |
| "loss": 2.4743, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "eval_loss": 4.076536655426025, | |
| "eval_runtime": 5.5106, | |
| "eval_samples_per_second": 77.85, | |
| "eval_steps_per_second": 39.016, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "learning_rate": 1.3671946710025964e-05, | |
| "loss": 2.4692, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "eval_loss": 4.073933124542236, | |
| "eval_runtime": 5.4811, | |
| "eval_samples_per_second": 78.269, | |
| "eval_steps_per_second": 39.226, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "learning_rate": 1.327895937688495e-05, | |
| "loss": 2.4651, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "eval_loss": 4.068994522094727, | |
| "eval_runtime": 5.4881, | |
| "eval_samples_per_second": 78.168, | |
| "eval_steps_per_second": 39.175, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "learning_rate": 1.2885578663030082e-05, | |
| "loss": 2.4885, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "eval_loss": 4.072272300720215, | |
| "eval_runtime": 5.4901, | |
| "eval_samples_per_second": 78.141, | |
| "eval_steps_per_second": 39.162, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "learning_rate": 1.2492591329889067e-05, | |
| "loss": 2.5023, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "eval_loss": 4.06748104095459, | |
| "eval_runtime": 5.5201, | |
| "eval_samples_per_second": 77.716, | |
| "eval_steps_per_second": 38.948, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 1.2099210616034198e-05, | |
| "loss": 2.4651, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "eval_loss": 4.064865589141846, | |
| "eval_runtime": 5.5135, | |
| "eval_samples_per_second": 77.809, | |
| "eval_steps_per_second": 38.995, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "learning_rate": 1.1706223282893185e-05, | |
| "loss": 2.4774, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "eval_loss": 4.06947660446167, | |
| "eval_runtime": 5.4995, | |
| "eval_samples_per_second": 78.006, | |
| "eval_steps_per_second": 39.094, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "learning_rate": 1.1312842569038316e-05, | |
| "loss": 2.4717, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "eval_loss": 4.055931091308594, | |
| "eval_runtime": 5.4978, | |
| "eval_samples_per_second": 78.032, | |
| "eval_steps_per_second": 39.107, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 1.0919461855183447e-05, | |
| "loss": 2.4856, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "eval_loss": 4.051236629486084, | |
| "eval_runtime": 5.5101, | |
| "eval_samples_per_second": 77.857, | |
| "eval_steps_per_second": 39.019, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "learning_rate": 1.0526474522042433e-05, | |
| "loss": 2.4572, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "eval_loss": 4.0473432540893555, | |
| "eval_runtime": 5.5015, | |
| "eval_samples_per_second": 77.979, | |
| "eval_steps_per_second": 39.08, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "learning_rate": 1.0133093808187564e-05, | |
| "loss": 2.486, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "eval_loss": 4.043802261352539, | |
| "eval_runtime": 5.52, | |
| "eval_samples_per_second": 77.717, | |
| "eval_steps_per_second": 38.949, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "learning_rate": 9.74010647504655e-06, | |
| "loss": 2.449, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "eval_loss": 4.038473606109619, | |
| "eval_runtime": 5.5095, | |
| "eval_samples_per_second": 77.866, | |
| "eval_steps_per_second": 39.024, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "learning_rate": 9.34672576119168e-06, | |
| "loss": 2.456, | |
| "step": 61000 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "eval_loss": 4.035487651824951, | |
| "eval_runtime": 5.5078, | |
| "eval_samples_per_second": 77.889, | |
| "eval_steps_per_second": 39.035, | |
| "step": 61000 | |
| }, | |
| { | |
| "epoch": 2.93, | |
| "learning_rate": 8.953345047336813e-06, | |
| "loss": 2.4802, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 2.93, | |
| "eval_loss": 4.0377516746521, | |
| "eval_runtime": 5.5124, | |
| "eval_samples_per_second": 77.824, | |
| "eval_steps_per_second": 39.003, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "learning_rate": 8.559964333481944e-06, | |
| "loss": 2.4635, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "eval_loss": 4.03075647354126, | |
| "eval_runtime": 5.5127, | |
| "eval_samples_per_second": 77.82, | |
| "eval_steps_per_second": 39.001, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "learning_rate": 8.16697700034093e-06, | |
| "loss": 2.3742, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "eval_loss": 4.048806667327881, | |
| "eval_runtime": 5.5265, | |
| "eval_samples_per_second": 77.626, | |
| "eval_steps_per_second": 38.904, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "learning_rate": 7.77359628648606e-06, | |
| "loss": 2.2371, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "eval_loss": 4.057917594909668, | |
| "eval_runtime": 5.5266, | |
| "eval_samples_per_second": 77.625, | |
| "eval_steps_per_second": 38.903, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "learning_rate": 7.380608953345048e-06, | |
| "loss": 2.2496, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "eval_loss": 4.062958717346191, | |
| "eval_runtime": 5.4861, | |
| "eval_samples_per_second": 78.197, | |
| "eval_steps_per_second": 39.19, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 3.16, | |
| "learning_rate": 6.987228239490178e-06, | |
| "loss": 2.2758, | |
| "step": 67000 | |
| }, | |
| { | |
| "epoch": 3.16, | |
| "eval_loss": 4.05156135559082, | |
| "eval_runtime": 5.5178, | |
| "eval_samples_per_second": 77.748, | |
| "eval_steps_per_second": 38.965, | |
| "step": 67000 | |
| }, | |
| { | |
| "epoch": 3.21, | |
| "learning_rate": 6.594240906349165e-06, | |
| "loss": 2.2489, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 3.21, | |
| "eval_loss": 4.058535575866699, | |
| "eval_runtime": 5.4932, | |
| "eval_samples_per_second": 78.097, | |
| "eval_steps_per_second": 39.14, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 3.26, | |
| "learning_rate": 6.200860192494296e-06, | |
| "loss": 2.2374, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 3.26, | |
| "eval_loss": 4.071547031402588, | |
| "eval_runtime": 5.5054, | |
| "eval_samples_per_second": 77.923, | |
| "eval_steps_per_second": 39.052, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 3.3, | |
| "learning_rate": 5.807872859353283e-06, | |
| "loss": 2.2862, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 3.3, | |
| "eval_loss": 4.050685405731201, | |
| "eval_runtime": 5.492, | |
| "eval_samples_per_second": 78.114, | |
| "eval_steps_per_second": 39.148, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 3.35, | |
| "learning_rate": 5.414492145498413e-06, | |
| "loss": 2.2502, | |
| "step": 71000 | |
| }, | |
| { | |
| "epoch": 3.35, | |
| "eval_loss": 4.051216125488281, | |
| "eval_runtime": 5.5012, | |
| "eval_samples_per_second": 77.983, | |
| "eval_steps_per_second": 39.082, | |
| "step": 71000 | |
| }, | |
| { | |
| "epoch": 3.4, | |
| "learning_rate": 5.021504812357399e-06, | |
| "loss": 2.238, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 3.4, | |
| "eval_loss": 4.054477214813232, | |
| "eval_runtime": 5.4933, | |
| "eval_samples_per_second": 78.095, | |
| "eval_steps_per_second": 39.139, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 3.45, | |
| "learning_rate": 4.628517479216386e-06, | |
| "loss": 2.2407, | |
| "step": 73000 | |
| }, | |
| { | |
| "epoch": 3.45, | |
| "eval_loss": 4.045898914337158, | |
| "eval_runtime": 5.483, | |
| "eval_samples_per_second": 78.241, | |
| "eval_steps_per_second": 39.212, | |
| "step": 73000 | |
| }, | |
| { | |
| "epoch": 3.49, | |
| "learning_rate": 4.235136765361517e-06, | |
| "loss": 2.2529, | |
| "step": 74000 | |
| }, | |
| { | |
| "epoch": 3.49, | |
| "eval_loss": 4.04516077041626, | |
| "eval_runtime": 5.4909, | |
| "eval_samples_per_second": 78.129, | |
| "eval_steps_per_second": 39.156, | |
| "step": 74000 | |
| }, | |
| { | |
| "epoch": 3.54, | |
| "learning_rate": 3.841756051506649e-06, | |
| "loss": 2.2453, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 3.54, | |
| "eval_loss": 4.045854091644287, | |
| "eval_runtime": 5.4906, | |
| "eval_samples_per_second": 78.134, | |
| "eval_steps_per_second": 39.158, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 3.59, | |
| "learning_rate": 3.4483753376517792e-06, | |
| "loss": 2.2314, | |
| "step": 76000 | |
| }, | |
| { | |
| "epoch": 3.59, | |
| "eval_loss": 4.041632175445557, | |
| "eval_runtime": 5.4892, | |
| "eval_samples_per_second": 78.153, | |
| "eval_steps_per_second": 39.168, | |
| "step": 76000 | |
| }, | |
| { | |
| "epoch": 3.63, | |
| "learning_rate": 3.0553880045107656e-06, | |
| "loss": 2.2408, | |
| "step": 77000 | |
| }, | |
| { | |
| "epoch": 3.63, | |
| "eval_loss": 4.037862300872803, | |
| "eval_runtime": 5.5221, | |
| "eval_samples_per_second": 77.687, | |
| "eval_steps_per_second": 38.934, | |
| "step": 77000 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "learning_rate": 2.662007290655897e-06, | |
| "loss": 2.2497, | |
| "step": 78000 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "eval_loss": 4.034841060638428, | |
| "eval_runtime": 5.5069, | |
| "eval_samples_per_second": 77.903, | |
| "eval_steps_per_second": 39.042, | |
| "step": 78000 | |
| }, | |
| { | |
| "epoch": 3.73, | |
| "learning_rate": 2.269413338228738e-06, | |
| "loss": 2.2475, | |
| "step": 79000 | |
| }, | |
| { | |
| "epoch": 3.73, | |
| "eval_loss": 4.037409782409668, | |
| "eval_runtime": 5.525, | |
| "eval_samples_per_second": 77.647, | |
| "eval_steps_per_second": 38.914, | |
| "step": 79000 | |
| }, | |
| { | |
| "epoch": 3.78, | |
| "learning_rate": 1.876032624373869e-06, | |
| "loss": 2.2376, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 3.78, | |
| "eval_loss": 4.03186559677124, | |
| "eval_runtime": 5.5186, | |
| "eval_samples_per_second": 77.737, | |
| "eval_steps_per_second": 38.959, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 3.82, | |
| "learning_rate": 1.4826519105190004e-06, | |
| "loss": 2.244, | |
| "step": 81000 | |
| }, | |
| { | |
| "epoch": 3.82, | |
| "eval_loss": 4.033052921295166, | |
| "eval_runtime": 5.5134, | |
| "eval_samples_per_second": 77.81, | |
| "eval_steps_per_second": 38.996, | |
| "step": 81000 | |
| }, | |
| { | |
| "epoch": 3.87, | |
| "learning_rate": 1.0892711966641317e-06, | |
| "loss": 2.2611, | |
| "step": 82000 | |
| }, | |
| { | |
| "epoch": 3.87, | |
| "eval_loss": 4.030561923980713, | |
| "eval_runtime": 5.5018, | |
| "eval_samples_per_second": 77.974, | |
| "eval_steps_per_second": 39.078, | |
| "step": 82000 | |
| }, | |
| { | |
| "epoch": 3.92, | |
| "learning_rate": 6.962838635231176e-07, | |
| "loss": 2.237, | |
| "step": 83000 | |
| }, | |
| { | |
| "epoch": 3.92, | |
| "eval_loss": 4.030078411102295, | |
| "eval_runtime": 5.5145, | |
| "eval_samples_per_second": 77.795, | |
| "eval_steps_per_second": 38.988, | |
| "step": 83000 | |
| }, | |
| { | |
| "epoch": 3.97, | |
| "learning_rate": 3.0290314966824894e-07, | |
| "loss": 2.2337, | |
| "step": 84000 | |
| }, | |
| { | |
| "epoch": 3.97, | |
| "eval_loss": 4.0290679931640625, | |
| "eval_runtime": 5.5016, | |
| "eval_samples_per_second": 77.978, | |
| "eval_steps_per_second": 39.08, | |
| "step": 84000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "step": 84736, | |
| "total_flos": 4.4280638078976e+16, | |
| "train_loss": 2.6336010373971255, | |
| "train_runtime": 9376.0618, | |
| "train_samples_per_second": 18.075, | |
| "train_steps_per_second": 9.037 | |
| } | |
| ], | |
| "max_steps": 84736, | |
| "num_train_epochs": 4, | |
| "total_flos": 4.4280638078976e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |